/v/ - Video Games

Vidya Gaems

Index Catalog Archive Bottom Refresh
Mode: Reply
Options
Subject
Message

Max message length: 8000

Files

Max file size: 32.00 MB

Max files: 5

Supported file types: GIF, JPG, PNG, WebM, OGG, and more

Captcha
E-mail
Password

(used to delete files and postings)

Misc

Remember to follow the rules

The backup domain is located at 8chan.se. .cc is a third fallback. TOR access can be found here, or you can access the TOR portal from the clearnet at Redchannit 2.0.



8chan.moe is a hobby project with no affiliation whatsoever to the administration of any other "8chan" site, past or present.

(37.04 KB 756x221 Manjaro Lel.png)
(415.58 KB 595x842 Linux shotabaiting.jpg)
Average Linux Gaming Thread: Manjaro Is Still A Meme Edition Anonymous 04/15/2021 (Thu) 16:46:42 Id:e77631 No. 285679
Didn't see one available, so here we go. Copy, Paste >What distro are you using? >Specs >Games you recently played >Wine or virtual machine? Other stuff >I'm interested in trying GNU/Linux, what should I start with? Linux Mint or Pop Os. These usually work with any computer you throw them at, if that computer meets the system requirements of course. This website should help with some more information: https://wiki.installgentoo.com/wiki/Babbies_First_Linux. >What are some interesting Linux operating systems to try at this time? If you want to get a version of Arch Linux up and running in as little time as possible, I think Endeavour Os is a good solution. Plus, it lets you choose what desktop environment to use during installation, if you have internet connection.
Taking linuxfag to another level
What should I tell my friend, who always tells me to play games that don´t work on Linux with him? Stuff like Destiny 2, Valorant, Apex, Halo MCC is kinda wonky on the coop...
>>285679 >What distro are you using? Artix Linux (based on Arch) (yes it does have GUI installers) >Games you recently played Quake(spasm) and (GZ)Doom >Wine or virtual machine? Neither Gaming on Lignux is not going to happen. All these half-assed attempts are doing is baiting people with false hopes and giving them nothing but disappointment. Just adapt to that like this: >install Wangblows on the HDD >install (your my prefered) Lignux distro on an external HDD >for everything non-vidya, boot into Lignux >for vidya boot into Wangblows but unplug your ethernet cable before it has a chance to do it's thing (phone home & shove updates up your ass) Also, remember to always separate your /home partion during installation, since that way if/when you have to reinstall the system, your data can be mercifully untouched
>What distro are you using? Ubuntu >Specs Toaster machine with i7 3770K CPU and RX 560 >Games you recently played GMod and City of Heroes >Wine or virtual machine? Wine. Currently I'm figuring out how the fug I need to build a car in GMod in a way where the wheels doesn't fucking spazz out on me so that I can later mount a light gun on it and kill some creatures, once I'm more familiar with vehicle constructing I will try to figure out how I can rip off the Harkonnen assault tank from Emperor battle for dune and make a turreted version because I like this design.
>>285759 Set your front wheels to have lower friction and assign them inversely to turning keys, set both back wheels to go forwards.
So I know there's the whole, Steam Runtime you can run everything (Including steam in) but what exactly is steam's development environment for linux? Asking since I have been having VR issues lately. Desktop view being upsidown and things like that.
>>285766 Those are not wired wheels, its the suspension variant which requires ballsocketing, ropes and hydraulics for it to work and of course I fucked it up once more by following this tutorial https://yewtu.be/watch?v=Flex-2so_Zs more closely.
>>285754 Been playing new and old games with wine and dxvk just fine, stop being baiting retard.
>>285679 >What distro are you using? Fedora the BOTNET linux >Specs Ryzen 3600 and Radeon 5600XT >Games you recently played mostly some old vns, gayming on linux sucks still >Wine or virtual machine? wine being a piratefag i still need wangblows to run recent games, currently dual booting windows 10 ltsc on separate ssd.
>>285716 Every single one of those games is irredeemable dog shit except for MCC, which is a re-release of older games on a superior platform. Just tell your friend to stop playing shit games. CSGO is the same thing as Valorant but it runs on Linux anyway. >>285754 >half-assed Anon I don't know if you've noticed but gaming on Linux is better than it has ever been. It isn't driven by the 0.001% of desktop autists like you and I who force square pegs into round holes. It's driven by a desire for platform independent software for ease of portability, as well as Valve trying to bypass Microsoft. We are just benefiting coincidentally. >>285781 Do you mean desktop environment? >VR on Linux Yeah good luck with that one.
>>285886 There it is there. Linux isn't viable as a standalone gaming box. We pirates have no choice but to take a pozz load down the throat from (((micro$0ft))).
>>285886 Have you considered not playing modern shitware with radically invasive "anti cheat" software?
>>285936 Linux is totally viable, just don't play esports shooters.
>>285917 I was more talking like, their entire system, I know up until ubuntu decide lol lets kill 32bit software that was their preferred OS, and maybe package setup. Also, when it works. VR on linux actually works pretty good. Honestly I thought it was totally busted for the longest time too until I found out, updating your headset base stations and controllers actually does something. Also Pavlov has shitty anticheat that explicitly LIES in logs about what causes it to "crash" on linux.
>>285679 > What distro are you using? Void Linux, it´s like Arch but I don´t need my mom to cancel all my appointments every few weeks. > Games you recently played Shadow Tactics, Freedom Planet > Wine or virtual machine? Wine, DOSBox and native. Virtual Machines are just too much hassle and you need a Windows copy. Wine works with pretty much everything I care about. >>285867 > Been playing new and old games with wine and dxvk just fine, stop being baiting retard. Retards who complain about Linux gaming always do so by comparing how well Windows software runs on the two systems. Yeah no shit, it´s going to run better on the OS it was written for. Don´t get me wrong, I am very appreciative of wine and dxvk and for older games Wine is great, but running Windows software will always be more or less of a hassle compared to native software. That´s not the fault of the OS though, Linux gaming itself if fine. It´s the game developers who refuse to support it and we have to rely on hacks and compatibility layers. Still better than running Windows though. >>285948 > radically invasive "anti cheat" software? I am still baffled that that shit is even legal. It is downright malware that fucks with your system and literally spies on you.
>>285754 Gaming on Linux is fine. >but only 80% of games work dude 95% of games worth playing work, since 95% of games worth playing are ancient. It's not ideal that some games don't work, but hearing "it's never going to happen" when I've played games on linux for years is hilarious. I'll be playing Dragon's Dogma tonight, suck my penis homo.
>>285679 >Wine or virtual machine? If you're a new user try to live without the gpu passthrough meme for 6 months, maybe a year so it doesn't become a crutch instead of learning to mess with wine and messing with it teaches you a lot about Linux never mind just wine itself. After that it's necessary for certain games to get acceptable performance or rarely to run at all. Waiting should be easier now (((Steam))) comes with easymode proton built in. >>285754 Dual booting leads the average user to spend more and more of their day in Windows as everything works at least passably there and rebooting is a hassle. The GPU passthrough meme is where fags should really be looking, it'll cost you a second GPU and some hassle but it's pretty easy to do following any one of a number of existing tutorials. >>285966 >you need a Windows copy Grab win10's iso from microsoft, skip registration during installation and if it's a newer verison that tries to force a microsoft account pull the network connection and it'll let you skip it then use one of the workarounds for running it without a legit key. Search for KMS smart activation script abbodi, top result that I'm not going to direct link is a retard friendly way to do this. >>285971 >95% of games worth playing work, since 95% of games worth playing are ancient. It's not ideal that some games don't work, but hearing "it's never going to happen" when I've played games on linux for years is hilarious. I'll be playing Dragon's Dogma tonight, suck my penis homo. Depending what you want to do this isn't always true. For example Relic's RTS games will run just fine but multiplayer is not viable because rather than properly syncing players they just sort of send events to all players and assume they will happen the same on each machine: running with wine fucks with the way the engine calculates things and dsyncs happen almost immediately. I know for a fact this affects CoH and DoW2 at least, not tried DoW yet but I can't imagine it's any different. Often multiplayer is a sore point also when a game's original online has died and the community workaround relies on some complicated automatic installation and subversion of the LAN functionality of the game. That said there are only a handful of games I've not been able to run at all then a few more that I can't play online, though some took a few hours of fucking around with to get going. >>285716 >Stuff like Destiny 2, Valorant, Apex, I'm pretty sure this is down to spyware-tier DRM meaning even a passthrough VM won't work. >Halo MCC is kinda wonky on the coop.. Have you checked wineHQ? To be fair MCC is wonky in general.
How practical is VM gaming? Booting back and forth looks mega fucking gay. What kind of setup do I need? Is the GPU passthrough thing necessary, and does the GPU need to have drivers available for the specific version of Windows you're virtualizing?
>>285679 >What distro are you using? Pop Os. My major criticism about it is the fact the developers chose to use Gnome over KDE or XFCE, desktop environments that are much more customizable and use use less ram. >Games you recently played Mostly emulation. Shining Force, Mario World. Thinking of also trying out some Ape Escape. >Wine or vm? Wine and proton.
>>285956 >VR on linux actually works pretty good. I guess that doesn't surprise me too much. Linux seems to have no shortage of driver autists that write drivers for everything so that they work natively. A stark contrast to the past, from what I've heard. However the problem is often getting higher level software to recognize those drivers, ironically. >>285979 >GPU passthrough I tried to set up a VM for gaming once and made the crucial mistake of assuming the GeForce 1080 it was a Christmas present from a friend, not my choice would work. Turns out Nvidia are turbojews who lie about their drivers containing a "bug" which prevents GeForce cards from working on VMs because they want you to buy the outrageously overpriced Quattro cards instead. This mysterious bug, according to my research, has had numerous successful workarounds over the past decade only for them to mysteriously stop working and new workarounds are found. I tried everything documented and nothing worked. I even copied the BIOS off the GPU and loaded it into KVM, nothing. Fuck Jewvidya. In short if you really want to make a gaming VM, make sure you buy AMD or do your research. >>285986 Dual booting is indeed mega fucking gay. I tried it for the better part of a year and ended up just going back to Win7 because closing all your shit just to play a game was insufferable. Then I upgraded my CPU and it was on an upgraded architecture unsupported by Win7. I didn't want random bluescreens so I went full Linux instead, especially because Win10 pisses me off constantly on my day job. GPU passthrough is absolutely necessary if you want decent performance, you'll need a secondary GPU for your host machine (i.e. the Linux OS so it can continue running in the background if you want to switch) but it can be a cheap shitty one, as long as it handles whatever desktop environment you want to use. >and does the GPU need to have drivers available for the specific version of Windows you're virtualizing? Well, yes. It's just as if you were installing the GPU into a Windows PC. However I'd be amazed if you couldn't find the drivers you need, assuming you're trying to use Win7. Just make sure you don't go with Barmitzvidia.
>>285996 >Turns out Nvidia are turbojews who lie about their drivers containing a "bug" which prevents GeForce cards from working on VMs because they want you to buy the outrageously overpriced Quattro cards instead. This mysterious bug, according to my research, has had numerous successful workarounds over the past decade only for them to mysteriously stop working and new workarounds are found. I tried everything documented and nothing worked. I even copied the BIOS off the GPU and loaded it into KVM, nothing. Fuck Jewvidya. I believe they actually just recently "fixed" the bug and released a VM compatible driver, believe it or not.
>>286013 I'm just going to assume for the sake of my own sanity that you are mistaken, because fuck those motherfuckers. I'm not setting up that shit again. Not in the near future anyway Besides Wine compatibility is improving faster than it used to. >>286017 God damnit anon.
>>285979 Does proton work with pirated games? If not, what fucking good is it?
>>286062 Proton is just Wine with some scripts layered on top. You can use it with pirated or non-Steam games if you want to. There are probably some fags who will come in and say that there isn't any point to using it unless you want to use Steam, but they are mistaken. There just usually isn't much of a reason to if there's a working guide on setting the game up in Wine. It might save you a few steps sometimes. My memory is too poor to remember the exact games I tried to run using Proton vs. normal Wine, but I seem to remember some of them having an easier time on Proton. >What fucking good is it? It's a thing Valve is putting money into. That's it, really.
>>286070 Valve putting money into proton puts money into wine, and valve is putting money into developing for the linux kernel as well. The success of valve and proton helps contribute to the ecosystem in general which is an overall good thing. It's unfortunate companies like CDPR aren't supporting linux as hardcore with GoG since I'd much rather also be getting support for DRM free content directly as well.
(24.21 MB 853x480 gens3_2.webm)
>>285679 >What distro are you using? I'm a Debianfag. Looking forward to finally updating to current year + 7 video drivers later this year. >Games you recently played Chrome, a game developed by Techland in the early 2000s. I unfortunately had to ditch it because of finding not one but several game breaking bugs. It was looking pretty bad anyway but I went through a great deal of effort to try to get it to work. Also played a bit of Sonic Generations. >Wine or virtual machine? Wine, but virtual machine sometimes to run installers and extractors if they won't work with wine. >>285754 >All these half-assed attempts are doing is baiting people with false hopes and giving them nothing but disappointment. Are you retarded?
>>286078 It's true that it's helped, since developers like GloriousEggroll cross over into doing lots of good work for Wine in general. He was working on making EasyAntiCheat work last year until he got sidetracked by writing the Media Foundations API. Not sure where he is now on that, but I was skeptical EAC would get working any time soon.
>>286062 Proton is more or less just fancy automatic wine so yes, it will. Really it's best for getting people over the shock of the initial few months of swapping from Windows though and presumably if you haven't made that hop yet you're probably a good goy with a large Steam library. >>285996 >I tried to set up a VM for gaming once and made the crucial mistake of assuming the GeForce 1080 it was a Christmas present from a friend, not my choice would work. Turns out Nvidia are turbojews who lie about their drivers containing a "bug" which prevents GeForce cards from working on VMs because they want you to buy the outrageously overpriced Quattro cards instead. This mysterious bug, according to my research, has had numerous successful workarounds over the past decade only for them to mysteriously stop working and new workarounds are found. Yes it's a known issue with fairly simple workarounds, but more generally Nvidia is a pain with Linux. >I tried everything documented and nothing worked. I even copied the BIOS off the GPU and loaded it into KVM, nothing. Fuck Jewvidya. Might also be to do with which GPU was being treated as the boot GPU, that can cause weirdness. There are workarounds for that as well including messing with UEFI settings and simply just swapping which slot the GPUs are in alongside some weird grub setting that's to do with how it displays the initial bootup I think. >>285986 >How practical is VM gaming? Booting back and forth looks mega fucking gay. VMs in and of themselves are too slow for practical vidya for anything but the oldest games. A WM with GPU passthrough however is extremely practical once it's set up: it can take between a few hours and be utterly painless or upwards of a few days of troubleshooting depending how lucky you get with hardware specific issues and how comfortable you are with the terminal. There are plenty of tutorials and almost certainly youtube e-celeb faggot tutorials you could follow that will spoonfeed you though. >What kind of setup do I need? Any recent CPU, two GPUs you can use an integrated one for the Linux host machine if you don't want to play serious vidya on it, but better to get two proper ones, a motherbord that's recent and actually has the right sized PCIe slots for both GPUs almost guaranteed if you don't cheap out, a non-meme case that can fit both GPUs, a sufficient PSU to handle running both GPUs again don't cheap out and you'll be fine and enough RAM to dedicate a good chunk to the guest OS without running into issues with the host (16GB total is plenty). In other words take any standard 'gaming' PC you've already got and add a second GPU and you're almost certainly fine. I'd also recommend having a dedicated SSD/HDD for the guest machine or a an SSD for its OS and a larger HDD to stick the games on and passing that through as a raw device too but that's not strictly needed. You'd also need either at least two monitors, one for each GPU, or more likely a physical switch which lets you plug two output cables from the two GPUs into one monitor and swap between them with a button. You can 'also use that switch to swap the keyboard and mouse between them but there are decent software solutions like Synergy/Barrier that are more convenient and remove the need for that: that's as seamless as moving your mouse between the Windows Monitor and the Linux one or hitting a hotkey to do so. >Is the GPU passthrough thing necessary For near native performance with the VM yes, but it's pretty simple to do following something like https://wiki.archlinux.org/index.php/PCI_passthrough_via_OVMF or similar (use Virt-Manager and follow that set of instructions, not the raw XML editing ones). And again you can actually play a lot of games on the Linux host without bothering with the VM at all using wine or if they have a native port. As I said unless you've got a specific addiction to a certain MP game you can and probably should go without the VM for the first six months so you probably adapt to the new way of doing things and don't use the VM as a crutch. >and does the GPU need to have drivers available for the specific version of Windows you're virtualizing? Once it's up and running you install the windows drivers from inside the VM just like you would on any windows machine (i.e. visit AMD or Nvidia's site and run the automatic tool). As far as Windows knows it's basically just seeing a normal GPU directly. So obviously your GPU will need to support Win7/Win10 but that will be almost guaranteed if it's even relatively recent. In fact that's how you install basically everything: it's just a virtual windows installation. Simplified somewhat but with passthrough what you're doing, more or less, is grabbing the second GPU at boot with a dummy driver on the (Linux) host and holding it so the host OS can't use it. Then when the VM guest (windows) is started up it gets passed that GPU so it can speak to it more directly instead of through the host OS. This compares to normal VM usage where sort of it has to run on a pretend GPU with a big performance hit. It looks scary and sounds complicated but honestly it's pretty simple to get yourself 90% of the way there, then if you hit problems in the last 10% ask in a thread like this or google a bunch and you'll be fine. After it's set up you're just interacting with an isolated Windows installation as if you'd booted up a Windows PC. Basically you end up treating Windows like a weird inbuilt console you can launch when you want to play specific vidya without the hassle of rebooting.
>>285986 >How practical is VM gaming? Not practical. VMs require a fuck load of time, effort and overall money investment to get anything running because running a VM is a huge resource hog so it's only useful if you got a lot of money to blow and want something that "just werkz" with things that typically wine/proton has issues with. It's a huge leap for something you could easily do by either having a 2nd pc with windows or simply dual booting. the fact that you don't have to dual boot isn't really that big of a reason to go for it. >>286062 It gives wine more sway in it's development along side with being about to finally sit at the negotiating table with the linux foundation. As it turns out Linux(kernel) devs are all jews who won't listen unless you're a multi-billion-dollar company >>286080 How much has that new kernel feature even done for the anti-cheat situation? I know that a good amount of wine devs have stated it's intended purpose was only just to circumvent DRM measures
>>286168 It worked for anti cheat at the time. However because breaking linux is specifically a feature, it no longer works.
>>286168 >Linux(kernel) devs are all jews Literally anyone can submit code retard.
>They don't know kkkkkkkkkkkkklk
>>286269 Don't know what? In a desperate bid to maintain the corporate network space microsoft is attempting to integrate themselves into linux? We all know that.
>>286271 They added DRM to Linux, and Rust code wwwwwwwwwwwwwwwwwwwwwww
>>286272 Are you talking about HDCP? Because that's not actually included unless specified for your distribution.
>>286273 Syscall User Dispatch, it's as diabolical as it reads. There's more stuff coming for 5.12 WOHORHORHORHORHOR Also, FSF is shambles https://gnu.tools/
Before I take the leap into Pop and encounter surprises, how are the peripheral compatibility on linux in general? Is it painless plug and play, or how much fiddling is involved. Will 360 or bone controllers work?
>>286278 Wired should be fine, but I do know that the xbone controller deliberately refuses bluetooth connections
>>286278 >Will 360 or bone controllers work? Yes.
>>286275 >i am a retard and don't understand what a syscall is cool.
>>285788 Hmm I've found a couple suspension tool for this game, I'm gonna try those damn tools out and hopefully maybe it will fix this damn issue, I've found also other tools for placing props lets see if with those are just as good to align constructs that way.
>>286258 >Literally anyone can submit code retard. yeah, you can submit it, through a poll request which will get denied every single time you try to push it, then they'll up merge their branch to a newer release that the internal devs work on then you got to update your branch for parity and then the cycle continues. Not to mention if you've had any "problematic" opinions in the past you're forbidden from merging indefinitely. To get any new features into the kernel it takes a fucking metric tone of bureaucracy and a dedicated team to back it. wine devs, work primarily with reverse engineering windows applications and they've been attempting to get shit working proper for a long time, only for them to get shut down time and time again. It literally took valve's direct involvement to get linux maintainers to pull any sort of useful feature. Linux foundation would rather put resources and development towards stupid shit suggested by board members rather then useful things. their priorities right now look like : 1) suck microsoft's dick 2) andriod andriod andriod 3) money 4) "year of the linux desktop" speaches where the ceo of the linux foundation is using a macbook. 5) let redhat/freedesktop do all the work
>>286278 >how are the peripheral compatibility on linux in general? As far as I know almost everything that requires a proprietary driver will most likely not work or have reduced functionality, e.g. fancy mouse and keyboard features are out although basic functionality will work because it's based on standardized protocols (USB HID class). >Will 360 or bone controllers work? I have a wired controller and it worked out of the box, even in wine. >>286302 >4) "year of the linux desktop" speaches where the ceo of the linux foundation is using a macbook. I want to murder someone.
>>286278 It is plug and play. I use DS4s and they work great on both wired and Bluetooth. >>286302 >4) "year of the linux desktop" speaches where the ceo of the linux foundation is using a macbook. It's a Unix-derived OS so it counts haha
>>286302 Don't forget 6) Mass appealing to women and other minorities that are too stupid to code efficiently.
I managed at least to build a gun turret, the wire wheel really does some wonders. Praise upon monolith for letting me build a functional turret! Now I just need to find better methods to build a manned gun turret as I need a bigger cannon as the other mutant subjects has a much greater constitution. >>286272 How much rust shit is added in now?
>>286360 NOOO YOU CAN'T BE OKAY WITH MACBOOKS EAT MY LINUX BLACKPILL REEEEE
(16.15 KB 280x280 oWo.jpg)
>>285716 Inform your friend that you want to be more than video game friends. Compliment the sound of their voice. Reminisce about their funny mannerisms and jokes. Agree to meet up later at their place. You show up and before any words are said: >Steal kiss. >Fondle their dongle. >Increased speed, whoa. >Make sticky milk flow. Target distracted, but clearly not tired enough. Drop your dignity and place their joystick between your thighs. Slather any leftover gamer fuel between your legs. Really work them out until they overflow. >Even more gamer rope than before. >They came already, what the fuck?!! A Windows user, all locked down, pent up, as expected. Now you've done it. They're up against your ram slot ready to slam. Here's your chance. Locate their desktop pc, take your Gentoo usb, and start the install.
>>285716 All of those games are fucking garbage, and even normalniggers are aware of how fucking insane the Valorant kernel anti-cheat is. I don't see why Halo MCC wouldn't work though, and it's the only one worth playing if you ignore the god awful matchmaking system and just play the campaign.
>>286687 Halo MCC doesn't work because 2 weeks before it came out they threw in Easy Anti Cheat and some other motherfuckery to deliberately break proton. It also broke modding which was a selling point
>>286855 The game lets you disable the anti-cheat at boot specifically to solve both those problems.
>>286857 Except a sideloaded second class version of the game with the majority of the community features disabled is NOT a solution to the issue. It's a shitty bandage for a shitty judge to point and say "Oh well they technically didn't break the law :^) case dismissed"
(579.71 KB 800x792 1368378094229.png)
>distro pic related >specs amd and good enough Been playing satisfactory, was kind of a pain in the ass to get running because I run steam in a chroot. >had to install some application called bubblewrap because proton uses it for containerizing games >bubblewrap does not like being run in a container (it considers a chroot a container) >required mounting the root of the chroot to the root of chroot??? game runs pretty good but to get it to display properly i've had to run it at a lower resolution because it doesn't seem allow the native resolution of my monitor maybe it's because it's in a container, maybe it's because I don't have a desktop environment it is so far the most processing intensive game I've run but I'm enjoying the conveyor belt autism >>286491 From what I've gathered, some lib for generating icons in desktop environments has made rust a hard dependency which has forced a lot of users to install the rust language when they update their system, otherwise they would have to completely remove everything that depends on that lib, which means breaking their desktop environment, and then blacklisting the lib if they can even do that on their system. I ran into the issue on gentoo and it's fucking ridiculous because rust has a fucking massive source to compile and can take DAYS on slower hardware (oh but it's okay goy, if you mask the source package it will just pull the binary instead! isn't that thoughtful?) The main issue is that rust is an immature language that is riddled with exploits and not any better than c++ or c# in terms of performance or functionality, so now I'm on the watch for any software I use to make sure they don't adopt this fucking meme language by completely re-writing their software in it making it mandatory to install this bloated fucking language.
(1.51 MB 1328x934 1618105580236.png)
>>286675 Great meme. >>287092 >Gentoo >Not even running a desktop environment Chad. I couldn't imagine myself only using a terminal or a command line. >The main issue is that rust is an immature language that is riddled with exploits and not any better than c++ or c# in terms of performance or functionality, so now I'm on the watch for any software I use to make sure they don't adopt this fucking meme language by completely re-writing their software in it making it mandatory to install this bloated fucking language. I've seen a lot of programming places pushing it lately. Some big companies have to be pushing it, I couldn't think of any other reason. I love C# so much, it's one of the things I'll actually give Microsoft credit for.
Oh monolith, I have failed to construct a vehicle that is worthy of usage, my first experiment of building a armored car failed terrible. It's all terrain capabilities is very poor, the mounted gun is highly unstable making it unsuitable for hit and run tactics and to make it worse the car is not stable even on flat ground which causes it to roll off all the damn time. >>287092 Damn that sounds fucked up, >rust becomes a hard dependency for some packages What the hell are those idiots thinking? >rust is riddled with exploits Got any examples? I thought that damn thing was supposed to be safer than c++ with its special unsafe markings or something.
>>287099 Much like a ton of places want you to know, only because it's new, and not because it's useful.
I like ghostbsd, it's comfy and I find it easier to install NVIDIA drivers better any linux kernel. All I did was sudo pkg install nvidia-driver-XXXXX then applied and it ended up working.
>Distro? Pop_os, thinking of slapping XFCE on it since after using GNOME for a few weeks, I realize I am not a fan. Looks nice but KDE does the same thing but better. I switch to Windows for certain games and programs. >specs RX590 (was wanting a 6800XT but the shortages, not paying the current prices) Ryzen 5600x 32GB of RAM 4TB of SSD with a 4 TB NAS server. >Games you've played? Final Fantasy 14, works great on Pop_OS with minimal tweaking. Proton is pretty good. >Wine or virtual machine? Wine all the way, VMs are more of a pain to deal with imo, only use them to test different configurations.
>>287099 >I couldn't imagine myself only using a terminal or a command line Not hard, man pages are your friend. Learn about piping and grep.
>>287111 If you do end up using KDE, personal advice. Thunar>Dolphin
>>287117 Thunar is my favorite file manager, so already ahead of you.
>>287111 >Pop_os, thinking of slapping XFCE on it since after using GNOME for a few weeks, I realize I am not a fan. Looks nice but KDE does the same thing but better. The sad thing about Pop is that it's tied with Gnome. Even if you install a different de, that gnome stuff will still be running and using RAM in the background. Removing Gnome will cause Pop to break, unless you know how to use it. XFCE, KDE, and cinnamon are much more customizable out of the box and use way less RAM. They should learn from Endeavour Os and let the user install whatever de they want want when installing the OS. >>287113 >Not hard, man pages are your friend. Learn about piping and grep. But I'm too lazy to do that! >>287117 >>287121 Thunar is sexy.
>>287123 Thunar Bulk file renamer for fixing casing issues wjth stuff designed for ntfs Good shit. for real though, I get on ntfs it doesn't cause problems but for fucks sake garrysmod server devs stop fucking up the casing of filenames for non workshop mods
>>287123 >Even if you install a different de, that gnome stuff will still be running and using RAM in the background Wait really? That is a huge shame, because other then GNOME I really like Pop_OS, and the company seems alright. Guess I'll consider using Debian again, but only if I can't fully switch over.
>>287128 >Wait really? Yeah. Installed kde-plasma-desktop and while using KDE, my system ended up using around 2 gigs of RAM with a couple tabs open in Firefox. However, that could have been on my end. Feel free to double check on your end with XFCE.
>>287100 I don't have examples off hand but here's an informative post on the rust meme: >>>/t/1295
>>287099 >I've seen a lot of programming places pushing it lately. Some big companies have to be pushing it, I couldn't think of any other reason. The logic behind the push for Rust is that C/++ in complex projects tends to lead to a lot of memory management related errors, true even with the most competent programmers and horribly so for incompetent ones or when messing around with multithreading, and that at best costs dev time and at worst can lead to security issues while Rust's inherent hard rules on ownership are supposed to cut down on these heavily. The ownership stuff that's supposed to guarantee memory safety is enforced at compile time so won't cause a performance overhead as you often run into with safer languages. So it sounds like paying the short-term cost of rewriting everything in Rust for the long term benefit of easier maintenance and less bugs and on paper this probably works. In practice however there are a few things that you'll run into with Rust. Firstly there's a lack of experienced programmers, and a bad programmer can make a mess with any language granted usually not one as bad as they'll make with C/++ but still. Secondly a lot of work ends up being done with 'unsafe' Rust which disables the memory safety features and thus the entire point for using the language. How much you have to use unsafe Rust to do certain things and how much it's a crutch used by the aforementioned bad developers is up for debate but it's common enough practice that they've had to document how to do it correctly. Thirdly, and frankly this is the largest problem for serious use of Rust, is that the team behind it doesn't really have a decent versioning system and constantly throws updates out hell they advertise this as a positive so the language will change constantly and often quite radically. This quickly becomes very annoying for active projects or for going back and working on an older project and will almost certainly cause maintenance issues down the line. This is a common trend in current software development term it Agile, continuous integration or whatever you want where instead of releasing stable versions with x years of given essential backporting/support they just throw the burden onto the downstream user because they think every project is like their little bespoke web front end with javascript horror and badly made database backend where constantly updating the libraries and languages used is acceptable and having users find errors in production is fine. This is unacceptable for a serious engineering project for obvious reasons of safety and security. Think of it, if you like, as stable release distros vs rolling release distros except for essential parts of your project and without an autist sitting there every day to fix things constantly. They do, on paper, group up a bunch of incremental changes into editions but this is very rare and doesn't really guarantee any sort of longer term support. Documentation also tends to lag behind the actual features until an edition is released and that's just an extra level of headache for serious use.
>>287335 They harp on very heavily about the memory safety, but that's not the only type of security issue too. It's quite easy to write a memory safe program that leaks data like a sieve because the developer was still retarded.
>>287335 That sounds like a nightmare. Thanks for filling me in.
(491.85 KB 1282x1050 fug23.png)
fug
>>287344 That is true also yes, which is part of what I meant by bad programmers making a mess regardless of language choice. The real issue with Rust that I was trying to get across is that it's not designed by engineers it's designed by web-focussed diva-developers which is why you get the techhipster style approach to documentation, releases and versioning and the 'it's the end user's problem' approach to long term support for anything because as I said they think everything is just like their web front end/simplistic but large scale database back end projects. This puts it in the same bucket as ideas like eventual consistency, agile development, devops and so on which are fine for hugely scaled up 'soft' problems (e.g. mass data collection for the likes of Google) where bugs have a minimal cost and the savings in performance or in development time from going cowboy matter more but are not practical for engineering problems where discovery-by-failure, shit documentation or being unable to guarantee how things will work for given versions is unacceptable see: banking, low-level embedded hardware and so on. >>287360 It's a shame really as there are some decent ideas behind the language but it is doomed to forever remain an immature language under its current structure. >>287363 Assuming that's a menu and the text is fucked you could try messing around with fonts and the like (winetricks corefonts for a start). Lots of vidya ends up with fucked up text unless you install fonts one way or another, though usually it's either missing or ugly rather than big cyan boxes, so just a random guess.
Gnome disks still pretty nice though I think I'm too retarded for Gparted
>>287335 >>287344 >>287369 I don´t get all the hype about Rust. Its big selling point is safety, but if I really cared that much about safety I would instead use Ada, which is literally a military-grade programming language commissioned by the US army. It is also used in safety-critical fields like aerospace and transportation, it´s old and tested and there is more than one implementation. I would really like to learn Ada, but the official book is like 1000 pages long and costs 100$. I know how to get a PDF, but I would still like a physical book more.
>>287378 >it´s old Downvoted!
(591.19 KB 1281x1049 damn123.png)
>>287369 Damn, I didn't think it'd be that easy! I Didn't encounter anything like this before so I didn't know it was even fixable and I was ready to give up since it's just a shitty mediocre game. I guess I'll have to actually give it a try now.
>>287378 I don't know much about Ada since it tends to be used in only its very specific fields and mostly governmental work. I know it has a reputation for being annoying to work with but that might just be the cost of safety. However the fact that it's old and untrendy as >>287386 says probably rules it out. >>287391 First thing I try with any font issue.
(24.06 KB 800x286 question.PNG)
>>287377 If your pc supports virtualization, use gparted in a VM and mess around. Pic related is from TestOut, my biggest gripe is the 1 to 1.5 times the RAM size for the swap file. I can imagine this being a issue if you have 32GB or more of RAM. That's a significant amount of space, hell 16GB swap files seems so. >>287378 If I had to choose a language it would be Python and C. Is C# any better then regular C?
>>287099 >C# I'd like to give C# a try, from what I've seen it looks like something I might like if I had to adopt a higher level language but the facts that it's associated with Microsoft and that to give it any kind of professional use I'd likely have to work for windowsfags are a huge turn off. Ultimately with Qt around I don't feel I'm missing out much in terms of functionality, outside of the web stuff maybe. >>287132 I don't think it's so easy to measure. I use cinnamon on my desktop and when I boot it's using around 1.5GB. My laptop has exactly the same setup and uses about 500mb. My desktop has 32GB and my laptop has 6GB, I'm guessing something's getting optimized or something. >>287335 >The logic behind the push for Rust is that C/++ in complex projects tends to lead to a lot of memory management related errors, true even with the most competent programmers How does that happen when you use RAII? >This quickly becomes very annoying for active projects or for going back and working on an older project and will almost certainly cause maintenance issues down the line. This is a common trend in current software development where instead of releasing stable versions with x years of given essential backporting/support they just throw the burden onto the downstream user because they think every project is like their little bespoke web front end with javascript horror and badly made database backend where constantly updating the libraries and languages used is acceptable and having users find errors in production is fine I hate shit shit so much. I delved a little into android development and I was just appalled by how contrived the whole thing is given the differences in android versions and how they behave and what things Google would allow or not. This whole thing started by developers trying to jew the end user while posing as engineers by doing things in such a retarded way that it demanded the user to shoulder computing, bandwidth, time or convenience cost and now it's just jews jewing other jews and the whole thing is retarded, yet people that use these technologies just accept it and categorically reject the idea of using older, stable, and proven technologies. >>287369 >web-focussed diva-developers What a sad phenomenon. I thought I was the only one that found people like that IRL. Then they look at you like if you were crazy when you tell them they're retarded.
C# is a wonderful language and is basically Java done right. If you're concerned about Microsoft, just use a different compiler and dont use Visual Studio
>>287363 Well, it seems this game's media has issues too. In-game cutscenes and dialogue won't play. On startup the shell displays an error message about gstreamer plugins missing for MPEG-1, and this might be related to the cutscenes because they're in the game's directory and they're indeed encoded in MPEG-1, however disabling winegstreamer has no effect which is weird. Ultimately there seems to be just 2 movies so I don't care too much about it. No idea about the in-game dialogue though and it also is not affected by disabling winegstreamer. The in-game dialogue is in a compressed archive in the game's directory and it seems it's just a bunch of WAV files so I don't know why they're giving wine problems or what to install. I'll poke around a bit more.
>>287132 >>287459 Tried running XFCE, only real gnome applicationr unning is gnome keyring for whatever reason. Otherwise XFCE works great.
>>287459 > Is C# any better then regular C? C# is nothing like C, it´s a slightly better Java (which is not saying much) but with a worse standard library. You will be writing pure OOP code and you will be at the mercy of wherever Microsoft takes the language. If you want something that performs similar to C but isn´t C then take a look at Nim: https://nim-lang.org/
How could I root an android phone on Linux? >>287516 >I'd like to give C# a try. Like >>287553 said, it's basically Java but good. If you don’t feel like running Rider by jetbrains, Monodevelop is a good enough solution. >I don't think it's so easy to measure. I use cinnamon on my desktop and when I boot it's using around 1.5GB. My laptop has exactly the same setup and uses about 500mb. My desktop has 32GB and my laptop has 6GB, I'm guessing something's getting optimized or something. That's really interesting. >>287612 I guess it's just on my end, KDE runs heavier on Pop, or I'm just a retard
>>287870 >How could I root an android phone on Linux? The same way you do it on Windows, you flash Magisk through recovery if I remember correctly.
>>287992 I get that, but when I get my Galaxy S7 Edge to its bootloader screen, fastboot can't find anything. My terminal gets stuck on "waiting for any device" or something like that.
Ah linux gaming threads, where people go to post /t/ posts on /v/.
>>287870 >advocating for C# in a Linux thread WTF. Isn't it proprietary Microsoft shit with dev tools (Visual Studio) only on Wangblowz.
>>288036 Mono is the open source version of C#, if that helps. Rider and Monodevelop are multi-platform ides for coding in C#/Mono.
>>287335 Didn't C++ already address that issue long ago with the addition of various kinds of smart pointers?
>>287516 RAII? What's that? Why don't you just use malloc/free? This is low level programming, didn't you know? Just go use Python if you don't want to manually manage every single resource yourself. If there's any memory issues, it's clearly the fault of the programming language and not the programmers. We don't make memory issues. We know what we're doing.
>>288036 They've been slowly open sourcing a bunch of stuff for some time. I don't know how people don't smell the blood in the water. >only on Wangblowz Never asked yourself what's that mono thing wine asks you to download? >>288054 I imagine they consider it not safe enough to the extent it warrants a completely new language and rewriting all the software, with a shift in management on the side.
>>288054 >>288112 But see, with smart pointers, you can still get the underlying pointer. And they're super serious excellent programmers who don't make mistakes, so clearly any problems relating to memory safety are the language's fault. There's other issues relating to memory safety which you can't get around via static analysis. Basically anything to do with data serialization in a non-retarded format.
>>287092 >From what I've gathered, some lib for generating icons in desktop environments has made rust a hard dependency which has forced a lot of users to install the rust language when they update their system, otherwise they would have to completely remove everything that depends on that lib, which means breaking their desktop environment, and then blacklisting the lib if they can even do that on their system. librsvg. It's a part of Gnome, and as the name implies, it handles SVG files. For some absolutely niggerific reason Gentoo maintainers make it a hard dependency for the base GTK3 lib because "muh circular dependencies". Never mind that compiling GTK3 separately without librsvg has no issues unless the programs need it explicitly. So we got four options: >Mask librsvg >=2.40 2.41 is where the tumor appeared. >Avoid everything GTK3/4-dependent >Compile everything GTK3/4-dependent outside of portage >Modify the ebuilds This (((RUST))) shit is getting pretty tiresome. >>288036 I feel like being a pedantic faggot so I'll point out that the proprietary MS shit is the .NET platform. That's not exactly the .NET-compatible languages like C# and F# themselves but rather the toolchains and the runtime environments where CIL (Common Intermediate Language - the intermediate language that C# and other .NET-compatible languages compile to. Think of it as analogous to Java bytecode.) is translated to native code. But nearly two decades ago a certain company (Ximian) started developing a FOSS implementation of the .NET runtime and accessories so that ideally you could use .NET programs and write C# code even on Linux/Unix. That implementation is called Mono. Incidentally, Unity Engine uses an old fork of Mono. Unity games have their actual game code in a few DLLs that can be decompiled with CIL-focused decompilers (ILSpy, dnSpy), and you can then use the Mono toolchain to recompile it back. Some modfags do that because Unity's implementation of Mono is stuck on an old version that doesn't even have asyncs. Reason: licensing faggotry between Unity and Xamarin (the company that was formed when Ximian got killed). By the way, Xamarin and by extension Mono are now MS subsidiaries, so MS has a lot of say in Mono's future.
(354.19 KB 500x655 i_have_given_up.png)
>>288768 I've stopped attempting to purge rust from my gentoo system. Every single application i've ever used for some reason has started pulling this dumb shit into my system and trying to purge it from a system I've already configured and installed everyone on is anoying as hell cause gentoo sees rust as a build dep not a normal dependency so it doesn't show up in equery. I still think rust is gay as fuck and i'll attempt to avoid what ever software i see use it. but when dep/libs start pulling it i can't fucking be assed. It's like looking for a needle in a hay stack, fuck that. i'm down with trying to purge it again if someone could make a concise list of libs/deps that pull rust and a list of forks/patches that remove the cancer I don't know why there doesn't exist a "anti-rust" overlay.
>>288768 i'd also like to mention that rust is literally the longest period to any fresh gentoo install, it would literally be in gentoo's best interest if the developers would start purging rust infected applications cause the already long install time got 100x slower with rust being pulled in. I had helped my brother install gentoo and it took i kid you not a whole 24+ hours to just compile rust alone. and he had a fucking threadripper
(174.15 KB 350x262 1377614479786.png)
>>288787 >threadripper >24+ hours to compile I believe it because rust is such a giant piece of shit, now imagine being forced to install it on a laptop with an quad core APU. I've got a better one yet, imagine trying to compile literally anything that relies on Qt and find out that you literally cant because the devs of Qt bundled their own build system called ninja (because it silently kills your update lel) that causes OOM-killer to kick in because ninja is preconfigured to use like 20 CORES and your makeopts (-jN+1) doesn't propagate to ninja so even on a good system you could experience OOM killing your update since it's allocating memory for cores you don't have which could be upwards of 1.5GB per core. One of my favorite browsers requires Qt so I can not update it unless I download the Qt source and patch it so that ninja isn't set to double the number of cores available, and you know what's a lot easier? Just not fucking updating ever again since it seems like I only lose things I want every time, like the xdm login manager has apparently been replaced, and I had mine riced out with custom login window and dual screen backgrounds but nobody else uses it so now it's "display-manager script" because it doesn't depend on x11-base/xorg-server, for this reason: >Its purpose is to provide the same startup mechanism for your chosen display manager (like GDM, SDDM etc. [2]) as xdm did previously, but without depending on x11-base/xorg-server. This is necessary to support new DMs that no longer depend on Xorg. >to support new DMs that no longer depend on Xorg Well I'm sorry gentoo team but I'm not USING those fucking DMs because I want to continue using XDM. I thought the whole point of gentoo was to build the system how "I" want, not conform to these arbitrary changes being made just to show there's "progress" even if it's not necessary.
(96.29 KB 791x330 loonix.jpg)
So Linuxfag anons, how is Wine performance these days? I tried running Windows games on Lunix through Wine about 4 or 5 years ago. It ran pretty much everything I threw at it without issues, but the problem is that performance was trash. Around 15% ~ 30% worse fps than running games native in Windows and more stuttering. For example, a game that ran at smooth 60 fps on Windows would run around 48 ~ 44 fps on Wine and with some bouts of stuttering. Of course, I'm not sure if the issue was with Wine itself, or video drivers or some other system service eating up resources, but I spent hours doing what I could to optimize everything, even tried different Wine builds, and only got slightly better results at best. I don't remember exactly the distro but it was either Ubuntu or Mint, and at the time I was using a GTX 970. I haven't touched Linux gaming ever since, but I've been wondering how things are now.
>>288880 You'd have to look up benchmark comparison videos on jewtube since I wouldn't know how much faster windows is. But I know there is a lot of work being done in the kernel to increase wine performance by valve, so it'll probably continue to improve. It's mostly just the fault of developers for using directx > vulkan though.
Dumb question,anyone else having problems opening the Vita's onboard storage? Seems to change the permissions on itself.
>>288880 As always with Wine: it depends. DXVK really helps a lot, it can make a game go from unplayable to pretty smooth, so that´s worth giving a try. If you want precise answers you have to do what >>288996 said and look into benchmarks.
>>288880 I play mostly older games so I don't notice any performance regression compared to wangblows, however games like Stalker still has choppy performance even with my RX 560 graphic card, I forget if I played it with DXVK on or not, I can't get Call of Chernobyl with Direct3D 10/11 to work for some reason as it crashes out on me with missing bump map texture. >>287157 Welp god damn I hope that doesn't mean I have to jump ship from linux too, I thought Lunix was supposed to be a decent alternative to wangblows.
>>288880 DXVK's first release was about 3 years ago so if you tried wine 4 years ago you weren't using it and performance would have probably sucked, which is confusing because you said you only had 15 to 30% less performance when I would expect more, but then again I wasn't using wine back then so I'm just guessing. With DXVK that difference should have been reduced significantly and it should have enabled better compatibility, but there's always going to be overhead and 10 to 20% doesn't sound crazy considering what's going on underneath and that 10% may be just 6 to 10 FPS.
Fugg I've downloaded this game called sexy beach premium resort 3 and I can't rebind the controls as it says I need a controller for that, is there any fix for that? I also cannot run the patched exe the magnet came with it.
>>288880 It really depends heavily on what libraries a game requires and how new the game is. Proton has been pushing compatibility with newer games and has pushed lots of funding and stability into wine as a result. The only real way to know how anything performs through wine is either test it yourself or check the wine application database. >>289068 To most users, they will not even care about the rust dependency, like if they're using arch/debian/redhat/etc their updates will just pull the binary for rust, they likely wont even realize it's been injected into their OS, just like when I was still using arch linux and one day I find that it has systemd infecting it.
>>289126 You could subvert the whole thing and directly remap through xkb. You dont even need to run a windows specific remap tool cause linux is actually quite liberal with keymaps. I don't know if there exists a gui application for xkb mapping though, so you may be out of luck there.
>>287516 >>288054 There are tools to improve the situation with C++ yes but with complicated projects abusing objects/blindly implementing OO patterns something that plagues C++ in general frankly, but it's still better than asking most people to work with C it still tends to get messy and even experienced devs can slip up here and there, particularly when working with a long-term codebase which has had many contributors and sloppy patches. Developers also end up subverting the functionality of RAII due to either overconfidence as >>288710 says or, with competent devs, because of expedience when tackling trickier problems. A particular area of pain comes with shared/weak pointers usually manifested with a fuckton of only shared pointers with a weak pointer being inserted here and there to 'fix' the mess created and unique pointers ignored as too difficult to use being overused then combined with either multi-threading or circular code (or both) to the point that the use of smart pointers becomes almost placebo. Some use of these is unavoidable even for competent programmers and thus issues can happen and not get caught particularly since testing is often just a cult nowadays instead of serious work Test Driven Development for example has been subverted from a good idea into a blind pattern, often combined with discarding dedicated testers and just having devs half-arse it. With Rust, on paper, you can't do something similar without being explicit in using unsafe Rust while when it's done with C++ it's often subtle and spread all over the codebase and so it can be harder to come in a few years down the line and diagnose what they've done. However as I said in practice so much of the 'serious' work in Rust projects ends up being done with unsafe Rust that it's not all that different. There are other reasons why even well written C++ using RAII can still run into issues for example whenever you have to roll your own RAII methods (e.g. for handling libraries that don't, or things other than simple memory allocation) rather than just using smart pointers (possibly to wrap those libraries). Once again on paper Rust shouldn't have this issue because its libraries are in the same language though Rust has many shittily made libraries itself and dependencies can quickly get annoying due to the aforementioned shit release cycle I think it can also natively use C libraries which obviously can't do this themselves which can cause issues when they're used with C++ too. There are further cases where RAII isn't as easy a solution as it sounds (another example: shared smart pointers come with a non-negligable overhead particularly during construction but also with reference counting) but it mostly revolves around having to have multiple objects involved in owning and possibly modifying the resource and I think you can trace a lot of that to the overuse of complicated objects. We can argue all day about firstly if that's ever necessary and secondly if, when complicated OO hierarchies are the goal, C/++ is the language you should be using (not that I'd advocate Rust for that instead) but that's where I see the issues occurring. But like I said the adoption of Rust by large companies and FOSS teams should be looked at from their perspective of having a codebase with potentially decades worth of (usually poorly documented) code where someone who has since left had to implement a clever workaround in the 3 hours he was assigned/before customers started screaming and then was pushed onto something else before he could clean it up or comment it (never mind document it more seriously). Then years down the line someone else likely a diversity hire or someone who only got the job by googling 'how to pass programming job interview' though even people looking at their own code can struggle sometimes has to deal with that when either it or something it interacts with is causing a problem/holding up a new feature and it takes days and days to even diagnose it never mind fix it. So they're sold on this idea of it being a language with very hard rules that should prevent those errors from occurring or, if unsafe Rust was used sparingly, would at least make it easy to locate the only parts of the codebase the problem could originate from. Maintenance is the problem of long term codebases not adding new features. Rust doesn't solve as many of these issues as it claims to, at least in the (admittedly limited) practical use I've seen of it and it adds its own maintenance burden with the shitty release and documentation practices I mentioned before which has already become quite noticeable despite how young it is. It's a nice idea given the growing complexity of codebases, the ever higher % of incompetents working as developers and particularly with multi-threading to have a language built for memory safety first then swapping explicitly - and not very often - to an unsafe mode rather than the other way around (through after-the-fact bolt on features) as C++ is but Rust is not suitable for serious use. Also I should mention I'm by no means an expert in either Rust or C/++ I'm just going off some of the horrors I've seen (and had to maintain) in actual enterprise codebases and from the occasional glance over some of the sloppier FOSS projects, so I'm sure other fags will have different opinions. And apologies on the shitty layout and poor explanation of these problems, it's hard to explain what it's like digging around in years worth of code that potentially hundreds of people of varying levels of competence have stuck their fingers into when people are demanding results in extremely short timeframes until you've experienced it. >>287516 >What a sad phenomenon. I thought I was the only one that found people like that IRL. Then they look at you like if you were crazy when you tell them they're retarded. They're all over tech companies nowadays. Retarded startup bosses hire them because they ace the interview and the standard whiteboard problems then they basically just implement whatever shitty design patterns they learnt in university or copied from previous employers/stackoverflow while treating anyone who tells them no as if they're just too old to understand modern programming as if having a fundamental knowledge of low-level code makes you somehow worse at your job. This is the reason so many smart devices have glaring security holes and so many things are now using https as a file transfer protocol or web browsers as a not actually consistent cross-platform basis for the GUI. >>287592 Sounds like it's having issues reading the files or the filetypes. You could try the simple solution of just uncompressing the dialogue files incase something's going wrong there or that was supposed to be done during the installation and that failed silently because of wine weirdness. After that check the wine output in the terminal or logs for clues.
>>287870 >>287516 >I don't think it's so easy to measure. I use cinnamon on my desktop and when I boot it's using around 1.5GB. My laptop has exactly the same setup and uses about 500mb. My desktop has 32GB and my laptop has 6GB, I'm guessing something's getting optimized or something. Cinnamon is pretty but inconsistent and, possibly depending what applets or setup you've got, will enjoy eating up variable amounts of memory as you leave it running. I don't want to just blindly call it a memory leak, as it seems to be setup dependent, but it's there. For this reason when I'm using it (or when I wasusing it) and want a better baseline, it's feeling sluggish or to play some vidya I lazily restart it. Amusingly this issue has recently been noted and the official fix is literally just to automate that restart once a given amount of RAM is used as per https://archive.ph/VYept#selection-1040.0-1040.1 (https://blog.linuxmint.com/?p=4037). On one specific (AMD GPU) machine it can use anything from 200MB to close to 1GB if settings are being messed with.You should really move to MATE if you want that style of DE and particularly if you're going to be playing vidya as as well as it being consistently lower on the resource usage you can manually disable compositing while Cinnamon will only disable compositing for fullscreen windows and so many native ports of Linux vidya are lazy borderless windows because of the fucky way X handles multiple monitors one big monitor and devs/porters not being able to handle the game starting top left. I believe you can, at least, nowadays disable global vsync on Cinnamon. I have swapped DEs on every machine barring the one in the office frankly I kept that mostly because others there use it and it makes it easier to teach them but also because it's fine for productivity for both of these reasons and again because MATE is functionally close enough to identical for my purposes. >>288880 Wine is fine for most older games and for a decent number of newer games. Unity games have consistently been a problem, such as there are any worth playing, but that's usually a case where windows users report low performance compared to their specs and the issue is exacerbated with wine. Always check WineHQ as a baseline as well, 9/10 times with an old game you can just blindly slap whatever it lists into a new wineprefix and things will just wurk.
>>289273 Well thanks for the help anyway, looks like I have to stop playing this game because it has piss poor performance which not even DXVK can fix it and it crashed on me once due to a thread fault, no fucking idea what is going on.
>>289337 >And apologies on the shitty layout and poor explanation of these problems Yeah, some parts could have been explained better. >so many things are now using https as a file transfer protocol HTTP is shit and I hate it but that's a difficult one to argue against unfortunately given that there's barely any other protocol you can use that's standardized and has good support. Even if there were, it's hard to argue against REST shit because even a 2 int nigger could understand how things are supposed to work. >web browsers as a consistent cross-platform basis for the GUI >take a peek into job board >c++ engineer for desktop app development position >mfw they wanted someone to implement a chromium-based browser to consume a javascript front end as their desktop app >You could try the simple solution of just uncompressing the dialogue files No, I think it's like with idtech3 engines where the game's files are compressed inside a .pk3 file or with idtech4 games where the files are inside a .pk4, it's just a container format to group and compress the game's files. Textures and music everything else is packed in the same way and that works fine. >After that check the wine output in the terminal or logs for clues. I did, but there was nothing. I added a few debug arguments to increase the verbosity of the different interfaces but there was no error, in fact the logs of the sound interface looked like the game was playing the sounds just fine and logged all the calls to the funcitons to play the WAV files. I installed a few libraries from winetricks like xaudio, dsound and such, but it also made no difference. It was interesting to experiment because the whole game worked great except for this but unfortunately I couldn't make it work. >>289344 >will enjoy eating up variable amounts of memory as you leave it running That's is not the issue here, I was talking about memory at startup, but in any case, If I leave it running for a day or so and close everything I have opened then the memory goes back roughly to the way it was when I booted it. Additionally, and perhaps I wasn't sufficiently clear in my original post, I was talking about system-wide memory usage (free -h) and not just cinnamon specifically. I've never really been able to explain why the memory usage is so high compared to my laptop. Perhaps there's some shady buffering going on in the dark due to the bigger monitors and the bigger video memory. I don't think adding the memory of all the running processes at boot would add up to 1.5GB, but ultimately I have more than enough ram so it doesn't bother me. >as per https://archive.ph/VYept#selection-1040.0-1040.1 >We know there still are a few leaks out there because we hear of people who come back to their computer after days of it being idle to find their Cinnamon process using 2GB, 4GB, 6GB of RAM My cinnamon process uses about 100 to 200MB of RAM and I don't think I've ever seen it go above that. >manually disable compositing I've never really bothered looking into the reasons why I'd want to do this. I've always played vidya just fine with things as they are. Similarly, I remember trying out several other DEs but Cinnamon was the best match for me for one reason or another so I have no reason to change it. It's not flawless but it's very comfortable.
(37.40 KB 500x280 ClipboardVideo.mp4)
IBM just declared Linux theirs: https://archive.is/BB5xD
(136.32 KB 1068x980 confused_takane.jpg)
>>289471 >You are not allowed to use your personal email account as a "hobby". You are an IBM employee 100% of the time. Since when company CEO niggers have the right to dictate what a person can or cannot do in their fucking spare time?
>>289472 Since they signed the contract. Companies are dicks like that, if you ever get hired on in a programing or artist job make sure to have a lawyer with you cause they will try to slip something like that in if you aren't watching carefully. This guy apparently didn't read the fine print and got a shit ass deal as a result.
>>289475 Good advice. Read what you get on paper and if fou dont understand shit let a lawyer figure it out.
>>289475 >Since they signed the contract. Contract law does not supercede rights, and with the exception of literal executive officers it's an explicitly defamatory case to make the premise that you are intentionally representing a company, when you are not presently working. Don't indulge contract faggotry Don't feed lawyers
>>289795 Yeah, seems like all he has to do is call their bluff. But chances are he doesn't want to displease his masters either way, so he will accept the abuse quietly even if he knows they are full of shit.
>>289471 Sorry for asking the obvious but that guy's work at IBM wasn't to work on VNIC right? Because otherwise this would make sense. >>289472 You'd be surprised. Some companies make you sign contracts that say everything you create or invent, even in your spare time, is property of theirs. The argument was something like that when you're working with them you're exposed to their technologies or ideas or other things that may influence you to make the invention in the first place so it belongs to them, if I remember correctly. I remember reading about this before so I think it's relatively well known. >>289795 >Contract law does not supercede rights You can sign many of your rights away, like for example the right of taking part in a class action lawsuit or even doing a lawsuit at all in exchange for (((arbitration))). This may just be the same case.
>>289941 >You'd be surprised. Some companies make you sign contracts that say everything you create or invent, even in your spare time, is property of theirs. Disney does this.
>>289941 >like for example the right of taking part in a class action lawsuit or even doing a lawsuit at all in exchange for (((arbitration))) Only the case if they foot the bill for administering arbitration. Otherwise it's null.
>>289943 >Disney does this. Is that why there is so much Disney porn? Frustrated employees wanting to vent and make Disney responsible for the shit they create? >>289941 >You can sign many of your rights away, like for example the right of taking part in a class action lawsuit or even doing a lawsuit at all in exchange for (((arbitration))). This may just be the same case. There are laws that even a contract cannot override. My favorite was the "by opening this CD envelope you agree to the term of use on the CD" that you would find on various CD envelopes. Entirely nonsense in most countries. Or those "warranty void if seal broken" stickers, breaking those cannot actually void your warranty.
>>289398 >Your added stuff about RAM I decided to do a casual "experiment" with the Pop KDE RAM. I compared it to how other distros use RAM at idle, give or take opening and closing an application or two. >Peppermint OS, which uses XFCE <Around 500mb at idle >Endeavour Os, with XFCE <Around 500mb at idle >Endeavour Os, with KDE <Let's say around 500-800mb at idle >Mint, Cinnamon <Around 600-800mb at idle >Pop Os, Gnome (the default de) <Around 1gig at idle, once it gets "settled in" >Pop Os, KDE <Around 1.2 gigs at idle So, from what I gathered, if you're autistic about RAM usage like me, and you want to use a specific DE with Pop, you might be better off with a distro that supports that DE out of the box. Feel free to conduct your own experiment to compare and contrast. In that case, what's your favorite DE, linuxfags? Mine is KDE, but I also like XFCE and Cinnamon. Also, this post has been a bit of a mess, but what are some good distributions that use kde support? I thought of using Kubuntu with a minimal installation and then working my way from there. >>290050 Yeah, I remember reading somewhere that Disney stores the porn, that their artists make out of spite, somewhere. But, that specific thing might be an urban legend.
>>289795 >>289935 >call their bluff >get fired >get blacklisted from job market >get defamed as a sex crazy nazi cultist racist pedo zoophile women hating baby kicker from russia >get banned on every speach platform >can't pay for legal fees cause jews Great idea.
>>290203 I'm surprised Mint can compete with Endeavour (which comes from Arch). On the other hand I'm surprised Pop managed to double Endeavor, lol. Can you do one for Debian with Cinnamon? I'm curious to about how it'll go but I think if I try it, it won't be a fair comparison given the differences in the system like I mentioned earlier. >you might be better off with a distro that supports that DE out of the box It might also be that KDE eats up more ram than XFCE or Gnome.
>>288880 As others have pointed out, you almost certainly were using wined3d - Wine's default Direct3D-to-OpenGL translator layer. Translation in and of itself is never zero cost, but the penalty depends on a variety of factors like how different the source and destination APIs are, how well the translator was programmed, and how well the source application you're trying to run was programmed. Imagine you're bilingual/multilingual and are trying to translate some text from your second language to your native one. If you're experienced at it, and the text is grammatically, syntactially, and stylistically fine; you'll have no issues, and the translation will be quick, quality, and fast enough to be realtime with minimal performance loss. If the text has many errors/typos, you might have to take brief pauses to parse and make sense of them, hampering translation speed. And if the text is some torrent of shit that makes bix nood ebonics look like high brow poetry, then you might just go "Fuck it! I give up!" and admit defeat. It's kinda the same with Wine and programs/games: if it's some pajeet-tier spaghetti mess that doesn't even know what the words "standards" and "specifications" are and relies on piles upon piles of undefined behaviour, it might run (albeit barely and with lotsa bugs) on Windows, but fail in Wine. Well programmed stuff tends to run in Wine OK. But then there's the OpenGL situation. In the late nineties OpenGL was seen as a viable competition to D3D. Carmack IIRC even preferred it. Then in 2000s OGL started losing steam and getting bloated. Things reached a retarded point in about mid 2010s when it became so turgid that some started ditching OpenGL in favor its mobile/embedded-oriented version OpenGL ES, which was less featureful but also leaner and with less runtime overhead. The embarassing part was that at the time D3D11 also had less overhead, so if you were trying to run Windows vidya in Wine, you had to lose performance twice: first on the translation, then because of the slower API. >I've been wondering how things are now. Some enterprising kraut created DXVK - a Direct3D-to-Vulkan translator that is not a part of Wine but is supposed to be used with it by being a drop-in replacement for wined3d. OpenGL's overhead concerns necessitated a radical solution, and Vulkan has been created to be just that - a closer to metal, simpler API where low runtime CPU overhead is an important goal. Consequently, running games through DXVK leads to a smaller performance penalty than running games through wined3d. How big the penalty is when compared to native Windows performance depends on a variety of factors (like game code quality, see above), but it's not uncommon for it to be negligible, and in some outlier cases there's even a sizable performance boost. Nier: Admirably Animated Android Ass is one such example. Given that it was the first game that the first public DXVK release officially supported (github.com/doitsujin/dxvk/releases/tag/v0.20), it's reasonable to infer that DXVK owes its existence to 2B's bountiful badonk. If you look through DXVK's github issues, you'll find bugreports even from Windows users. Apparently, running Direct3D11 games through DXVK on Windows can lead to better performance than running them there natively. Unexpected, but logical, since back in the day D3D11 also used to be criticized for bad CPU overhead (it just wasn't quite as bad as OpenGL's). That was mostly because AMD had poor as fuck driver level optimization for it due to being borderline bankrupt and couldn't spare a cent on it, and NVIDIA didn't bother because it only needed to do slightly better than AMD to stay in the spotlight. Now because the competition is healthier, they do pay attention to optimization, but the focus has shifted from D3D11 to D3D12 and Vulkan. So what performance Winfags running DXVK lose on translating D3D11 to Vulkan, they more than regain by running a faster Vulkan path. In some ways, this is applicable to Wine+DXVK on Linux too. >>289398 >>manually disable compositing >I've never really bothered looking into the reasons why I'd want to do this. I've always played vidya just fine with things as they are. If the game's running fullscreen (borderless or proper) and has its own means of enforcing Vsync, the compositor ends up being useless at that time and is simply yet another step in the pipeline. Best case scenario, you incur a tiny imperceptible latency hit; worst case scenario, it doesn't play nice with the game's syncing, and shit gets stuttery. I've heard horror stories involving Mutter doing that. >>289941 >>289943 >everything you create or invent, even in your spare time, is property of theirs. I remember Zenimax bitching at Carmack, claiming that his contributions to Oculus were their IP.
>>289398 >I've never really bothered looking into the reasons why I'd want to do this. I've always played vidya just fine with things as they are. I've used compton before, i can with 100% confidence that compositing is the devil when it comes to playing games on Xorg. Although most DEs (atleast the modern good ones) do have a method of automatically disabling compositing when fullscreened. manually turning off compositing really depends on the DE/WM you use. or just use wayland cause it gives both compositing and doesn't have any overhead. and Xwayland makes it actually usable as a daily driver
(260.84 KB 612x753 cinnamon edition.jpg)
>>290383 >Can you test Debian with Cinnamon? >lol Alright, but only because you're cute. I had to use a virtual machine, because I didn't have a spare machine or hard drive. Also, I used the unofficial (for extra firmware support), Debian stable, live cinnamon edition (which uses the calamares installer- because I was too lazy to install it the "normal debian" way). It used 700 something megabytes in RAM at idle. That's pretty comparable to Mint. If I had to choose between using Debian or Mint's implementation of Cinnamon, I'd go with or stick with Mint, since Mint is more convenient to use. >It might also be that KDE eats up more ram than XFCE or Gnome. KDE using up more RAM than Gnome is a hard pill that I'm not ready to swallow.
(98.17 KB 456x380 heh.jpg)
>>290527 >Nier: Admirably Animated Android Ass
Where were you when they actively exploited mainline? https://archive.is/m1Kod
>>290694 >A few minutes with anyone with the semblance of knowledge of C can see >that your submissions do NOT do anything at all, so to think that a tool >created them, and then that you thought they were a valid "fix" is >totally negligent on your part, not ours. You are the one at fault, it >is not our job to be the test subjects of a tool you create. >... >Because of this, I will now have to ban all future contributions from >your University and rip out your previous contributions, as they were >obviously submitted in bad-faith with the intent to cause problems. >- greg k-h
>>290694 And the people responsible for these patches are ///Qiushi Wu\\\ and ///Kangjie Lu\\\
>>290694 The dumbest part of this whole thing is the retarded patches actually got pushed to stable. and they tried to pull that bs "tee hee i don't know much about the kernel oopsie woopsie UwU sowwy" shit
(1.40 MB 900x1430 1570888883546.png)
>>290694 >>290721 >university of minnesota HAHAHAHA
i really want to know how deep these bugmen managed to nest this shit. Did this go as far as to be included in an actual build? they said it made it's way to stable...
>>290744 Didn't make it to any builds. Just another case of university niggers getting caught doing stupid shit and playing the victim.
>>290570 >I had to use a virtual machine That kind of defeats the purpose but I was still expecting less consumption. >>290694 Despite how retarded this is it's a good thing that it's come up and is being discussed. And that it was carried out in such a dumb manner that it was found out and exposed. >>290722 I thought actually getting code in the kernel was near impossible given how much it is controlled by established people and corporations.
>>290826 >I thought actually getting code in the kernel was near impossible given how much it is controlled by established people and corporations. It is, as an individual. This was a university abusing their position. Maybe i should've stated earlier that universities also get preferential treatment cause muh academics. But i guess that doesn't matter now cause these retards went as far as doing the most dumb shit for brains approach to sneaking exploits into the codebase. like what fucking idiot intentionally makes an exploit then writes a fucking published paper on the subject of their crime.
>>290847 >like what fucking idiot intentionally makes an exploit then writes a fucking published paper on the subject of their crime Chinks clearly.
>>290847 >like what fucking idiot intentionally makes an exploit then writes a fucking published paper on the subject of their crime. Academics have done far worse just to get published, if you don't get yourself published your career will go nowhere.
>>288996 >>289038 >>289068 >>289110 >>289256 Thanks for the replies anons. Yes, when I tried Wine, DXVK wasn't around yet. It translated DirectX to OpenGL. Like some of you say, Wine ran fine for pre-2010s dx9 games, but performance was shit in dx11. Newer games that offered a dx9 mode as an option would run much better in it (i.e. ETS2). As for Unity games, it doesn't surprise me that they run badly, Unity performance is terrible on Windows as well. Unity tends to be pretty taxing on CPU for some reason, and screen space reflection effects on Unity can bring even a RTX 2070 to its knees Unless the dev is competent and experienced enough to change so much code that it ends up becoming essentially a new custom engine only derived from Unity (like the Overload devs did). >>290527 Thanks for the detailed answer and history lesson. I know about DXVK but didn't know it made such a significant difference.
>>291496 >Unity performance is bad I use Unity to make games, and it's sad to see.
>>291496 >>291524 >Unity has shit performance Fugg, I hope Godot won't end up becoming a mene like Unity is already.
>>291546 Unity is still worth using in my opinion. Good games like Hollow Knight and Endless Legend were made in it. Plus, it lets you easily port to many systems, like Linux. Usually, the poor performance is due to the devs being inexperienced. The benefit, and downside to using Unity, somewhat similar to Godot, is that anyone with enough autism can use it to make games with.
>>291496 >Unity tends to be pretty taxing on CPU for some reason DXVK is a huge improvement in the GPU department but we forgot to mention that there have been improvements in other areas too. ESYNC was introduced to help with processor load in games and it can make a world of difference in games that are very processor intensive. Unfortunately I don't know when exactly it was introduced since it's a feature and not a separate project so it's a bit more difficult to check, but I imagine it wasn't that long ago. If I remember correctly there's even a different and more recent implementation of the underlying idea that yields even better results but when I checked it required you to recompile your kernel and since ESYNC+DXVK works amazingly well I didn't really look much more into it.
>>291553 Yet in practice barely any developers bother deploying Linux version of their games when using Unity. So not even wine DXVK will help offset that shitty performance. >is that anyone with enough autism can use it to make games with. I'm already on it.
>>290694 jesus fucking christ >>291546 the problem is that unity and other kits of that sort is like giving a tumblr "artist" an entire collection of art tools used by davinci, salvador dahli and bob ross or giving an entire workshop of with top quality wood working tools, chisels, saws, etc to someone who wants to make a bird house but never hammered a nail before the tools may work and work well in skilled hands but that collection of tools doesn't itself make you skilled, a bunch of devs can shit out indie games made in unity and as long as they work then they'll say "Good enough!" because they don't really lose anything by not optimizing it, there's no punishment for being bad at using the tools unless it's so bad they cant even run it on their own machine to which they're forced to fix it until it runs
>>290203 I prefer XFCE, KDE is my second choice with Cinnamon a close third. I do not like GNOME very much, unnecessary bloat and their default file manager is kinda shit. If I want that I'll just use KDE, least my shit doesn't break if I remove the bloat unlike GNOME (at least from my experience). Use LXQT for servers if I decide to use a GUI.
>>290570 > I used the unofficial (for extra firmware support), Debian stable Does this come with the non-free firmware? Pretty sure my ethernet adapter uses it, and default Debian doesn't have the drivers for it.
>>291677 (checked) >the problem is that unity and other kits of that sort is like giving a tumblr "artist" an entire collection of art tools used by davinci, salvador dahli and bob ross or giving an entire workshop of with top quality wood working tools, chisels, saws, etc to someone who wants to make a bird house but never hammered a nail before That is a fairly good comparison, and due to that "easy" of usage Unity is now forever stuck with leaving players a bad "after taste" because developers just have no fucking idea what it means using a full blown game engine for their lame demo games or what it means using a garbage collected language, I'm just going to pull it out of my ass and assume that one of the reasons why Unity games run like shit is people think that "Language is garbage collected = Never have to free up objects!", I looked up the profiler and one time I made a bug where projectils are not truly destroyed when they collided with something and the reason for that was that the function call was in wrong order, so I fixed this bug and noticed firing weapons uses up slightly less memory than it did before. >there's no punishment for being bad at using the tools unless it's so bad they cant even run it on their own machine to which they're forced to fix it until it runs Yeah it really shows, admittedly the only unity games I have played so far where some really crappy ones which looked barebones but idk I just wanted to check what that one stupid site had to offer whose URL I have forgotten. Those games in question barely had any shit going on and where so dumb with like only a handful of zombies/soldiers roaming around in the most jankiest way possible and that awesome action was more than enough to almost cripple my whole system to a halt. Yet funnily enough the Kerbal Space Program game didn't had lagged so badly, the performance was far from great but it wasn't exactly shit either.
>>291684 Did you add the non-free and contrib repos?
>>291689 I did not. Found the unoffical ISO. Will try it later.
>>291690 Firmware blobs are located in those repos. Without them you can't boot (at least some) AMD cards.
I remember a program that you can input in the terminal to test game controller vibration but I don't know what it was called. Could any anon fill me in?
>>292515 I don't know the program you're looking for specifically but in case you haven't found anything else you can test vibration directly through wine.
How's this build? https://www.csl-computer.com/k/0mm9tEhb >>292515 I just use steam or emulator pad GUIs for anything controller related.
>>297176 It's really expensive, kraut anon. However, knowing how much the crypto miners have jacked up the price, it might be a better deal than buying all the parts separately.
>>297275 jacked up the gpu prices*.
>>297275 Pre builts are cheaper than buying seperately atm for my country which isn't GER Aside from the GPU price, I just wanted to optimise the rest of the build. For performance and overall price. Wanted some more eyes on it before I commit to it. This 8 year old rig is getting annoying to use and I wanna sell it before it becomes junk. Probably the best price/time I'll ever sell it for now too considering the prices of retail and new.
>>297311 Samefag you replied to, in case my id gets changed. Fair enough. If you think it's the best deal you can find on the market, then go for it. I was in the same boat when I was building my first computer, where a prebuilt was more cost effective.
>>286017 wait >Do you need to have more than one GPU installed or can you leverage the same GPU being used by the host OS for virtualization? <One GPU is required for the Linux host OS and one GPU is required for the Windows virtual machine. so fucking noting?
>>297354 I don't understand your question.
>>297355 you still need 2 GPUs. I don't see the point
>>297356 That's how GPU passthrough has always worked. That article is about how nvidia's drivers now officially support GPU passthrough instead of actively breaking when they detect they're in a VM. There is work being done on SR-IOV-like shit with some hacked 3090 drivers but that's probably not going to be viable. Maybe intel's GPU's might be a solution in the next few years.
>>297357 oh ok. I guess I missunderstood
>>297356 I've seen a few guys make a script that releases the GPU from the host OS whenever a virtual machine is started up but that means you can't display both the Linux and Windows machine at the same time so the only benefits you get from dual booting is that it just takes a little less time to switch between the two.
>>297356 Needing 2 GPUs is horseshit.
>>292515 fftest maybe?
New kernel?
>>297786 The hope is someday that the entire graphics stack will be virtualised even on windows. That way it doesn't matter what OS you're running, games will run the same on everything. But for now, I wonder if iGPUs would work as the host graphics.
>>298073 >But for now, I wonder if iGPUs would work as the host graphics. They will indeed, though you might run into the hassle that is passing through the boot GPU. It's possible but annoying.
>>298086 > It's possible but annoying. I personally stopped bothering with that shit years ago. If a game doesn't work native or through wine/proton (which often require tweaks too) then I ain't playing it and I'm certainly not dual booting for fucking video games. I've got a windows laptop lent to me for work and I've only used it once because the web version of Word doesn't have the references feature. Either the industry (both hardware and software/games) smarten the fuck up and virtualise the graphics stack to be secure and OS-agnostic or I will continue to abstain from more than half of the AAA games and software market. And that's fine by me. Less shit video games in my life.
>>298093 My man.
>>297786 That's just a consequence of trying to achieve good performance from a VM, which is the whole point of doing passthrough, so that you can play windows native games or run other windows native software without having to lose <50% performance graphically. If you don't have 2 graphics cards to use for passthrough then you may as well dual boot. I was going to try doing gpu passthrough but then gpus became the new gold rush and I'm not going to pay a huge markup for one just to have the privilege, so if it doesn't work through wine or proton I just don't bother with it for now.
>>298093 >Either the industry (both hardware and software/games) smarten the fuck up and virtualise the graphics stack to be secure and OS-agnostic or I will continue to abstain from more than half of the AAA games and software market. And that's fine by me. Less shit video games in my life. That's not a bad approach honestly. That's mostly how I do it excepting a few multiplayer games windowsfag friends like to play that utterly refuse to run with Wine/Proton but frankly there's less and less of those worth caring about that still have actual populations playing every year. That's also because some older games with multiplayer fanpatch workarounds also don't play well with Wine even if their single-player portion is fine. Frankly I've gone months and months without booting up the Windows VM at all and not missed it.
>>298073 >But for now, I wonder if iGPUs would work as the host graphics. Yeah they work fine as a host GPU for passthrough.
>>298184 Does anyone know if DDDA's Online Mode works well through Wine/Lutris? I know there's an install script for the latter available, but I'm wondering if it allows online to work. The GOG version at least should work, right?
Gonna ask again, anybody having issues with trying to connect to the onboard storage of the psvita?
>>300259 Nope.
(349.32 KB 1920x1003 JingPad more like ChinkPad.png)
(257.26 KB 1889x903 JingLing About Page.png)
https://archive.is/878OG https://archive.ph/sB5Q8 https://archive.ph/d2INN Maybe this could get posted in one of the GamerGay threads, but it's most relevant here and /t/, but barely anyone posts there. A company is trying to make a tablet with an actual Linux OS, not a gimped version like Android. Even if you don't care about Linux, being able to have fully functional operating system in the portabiity of a tablet is nothing to scoff at. You could edit videos, draw, play games that aren't mobile garbage, and a whole bunch of other things. Sounds great, right? What's the downside? It's being made by a West Taiwanese company. But, it gets better. They're running a crowdfunding campaign for this thing.
ayy I've got myself 2 new hard/soft drives. Can I setup in lunix gparted program to make the new HDD as my primary /home directory and my SDD for dumping normal programs and as a /root directory somehow?
>>300493 Yes you can nest your HDD in your /home folder.
>>300228 I'm afraid I've not tried it. WineHQ reports the game itself runs fine but nobody seems to have tested the online play other than this https://appdb.winehq.org/objectManager.php?sClass=version&iId=33193 old result for the (((Steam))) version. As a general rule of thumb Wine works better for online stuff the more 'standard' it is so if you can run the game you'll probably be ok unless it's doing a ton of nip spaghetticode for the online functionality.
>>297427 I have that. it works fine, but it's not perfect. you're still forced to restart X, which is a very close thing to just rebooting.
>>300309 Linux on ARM is great, but not for gaming. No native games are compiled for it, you would be stuck with emulators and the occasional source port. And tablet SOC's usually have terrible graphics support, with only OpenGL ES support instead of full desktop GL, though Vulkan might be changing that.
(2.61 MB 2156x3040 smug_russia.jpg)
>>300309 >Beijing and California (but I repeat myself) Yeah, they can suck my Jing-a-Ling.
>>300507 Ah that's good, so then I can finally soon "retire" my old HDD to backup type.
>>300660 Of course that's only if you WANT you hard drive nested there, you can put it in a traditional mount point, and in some rare cases, for example, software that has a string limit for accessed data, you may find things working more, correctly, in a shortened /media/<nameofdrive> location. rather than /home/<youruser>/<yourdrive>
>>300309 >A company is trying to make a tablet with an actual Linux OS This just sounds like libreum/pyra 2.0 bets on this thing being way overprised, underpowered and slow as shit while also having gimped usability, because of shoddy maintainers having to maintain something that isn't x86. Not to mention completely pointless. unlike libreum which had a point: phones. This has no point what so ever. If you want to get a tablet that uses linux, just grab any recent tablet pc. since tablets have higher surface area they work quite well with modern x86 cpus so just throwing on linux of choice on one of those would completely out perform this shit. Why would you want to be locked into using chinkware when perfectly usable tablets exist already?
>>300541 Fair enough. Thanks, anon; that page gives me something to go on, at least.
>>300493 You can set up drives and mount points however you want really. There are some things I'd recommend, like only using the hdd for things that will perform a lot of writes, and documents/images, since ssd's have a set limit of how many writes to a cell it can take, and hdds should be less prone to failure these days. Also if you have enough memory, I recommend having /tmp set to a ramdisk instead of the sdd, or if you don't have enough memory then just use the hdd for /tmp since it will see a lot of writes. >HDD: /home >RAM: /tmp >SSD: everything else If you plan on using hibernate make sure the swap partition is double the amount of RAM you have, it needs at least enough space to store an image of everything in memory but more swap never hurts. Depending on your distro you may need to manually set a cron job for fstrim which is essential for ssds.
what is the best way to use the keyboard layers beyond four? example: "a" by itself is layer one, a+shift=A layer two, a+altgr=æ layer three, a+shift+altgr=Æ layer four, a+???=*something* layer five, etc etc...
>>300892 > since ssd's have a set limit of how many writes to a cell it can take I have a MLC SSD and according to other anon it would require 100GB of daily writes for it to wear down that fast. >Also if you have enough memory, I recommend having /tmp set to a ramdisk instead of the sdd, or if you don't have enough memory then just use the hdd for /tmp since it will see a lot of writes. Are you sure I should use the HDD for the cache file? I'm already using 2GB cache file on my HDD and everytime it gets utilized my computer becomes fucking slow. >If you plan on using hibernate make sure the swap partition is double the amount of RAM you have Is there a way in Nigbuntu to retain program states when it is shut down? On Linux Mint XFCE there was a setting that would let me do that and it kind of worked.
>>300892 That kills the point of swap. Either put it on SSD or get rid off swap altogether, having it on HDD is a nightmare. Also swap wont be ruining SSD if you configure the swap correctly, like this: xed admin:///etc/sysctl.conf vm.swappiness=20 Having a swap/pagefile on HDD always hurts, t. expert who has been removing it from his wangblows and loonix machines since 2005.
>>300966 >Are you sure I should use the HDD for the cache file? /tmp gets used by a lot of programs, and gets cleared on reboot, so there's more writes to /tmp than most other locations that's why I suggested using a ramdisk for it instead since personally I don't want to waste my ssd's write cycles with it, depending on what programs you run I don't see it causing a significant slowdown to put it on the hdd though >Is there a way in Nigbuntu to retain program states when it is shut down? It's probably different than how it's done on gentoo, but typically when you hibernate the image is stored on the swap partition so it pays to have more swap space than RAM, and the boot option to try the swap partition needs to be added to grub otherwise it will boot the default image and end up clearing the swap partition I'm pretty sure you can make another partition for hibernation images but I've never done that >>301019 /tmp and swap are not the same thing case in point, my /tmp partition is a tmpfs filesystem configured to use 16gb of RAM so on boot it sets up the /tmp directory in memory my swap partition is on an ssd and has 32gb allocated to it
>using ram for /tmp Wouldn't that defeat the entire point of having a filesystem based tempdir? I thought one of the biggest reasons developers use /tmp is because it doesn't have to be allocated to a heap like what using ram would require. Using /tmp you could store information way past the max capacity of your ram and not have to worry about memory management. And you'd be less prone to getting memory errors.. If the developer never intended to use ram in the first place wouldn't there be a reason for it? And it's not like the advantage of using ram tempdir would be worth it since the writes of modern SSDs are rather large that it'd take over a decade of constant use to even reach the max write.
>>301223 I'll give you an example of /tmp usage, when you run blender and render a scene, it saves frames of the render in a /tmp directory, so it has a significant payoff to have /tmp mounted as a tmpfs in RAM for that quicker write speed. There are also a number of applications that will write logs to /tmp, Steam is one of those applications. If you use pulseaudio it's writing data to /tmp, if you have systemd it's writing to tmp, some things even will write a crash dump to /tmp, your e-mail client is probably writing to /tmp. The fact of the matter is that /tmp sees a lot of traffic depending on what/how many applications you have running at any given time, and the fact that all this is temporary hence why it's being saved to /tmp and not to a more permanent location makes it that much more ideal to simply use memory as a /tmp directory. I have enough RAM that I lose nothing by utilizing it this way, and to argue that the 'developer' never intended to use ram in the first place is a rather ignorant way of looking at things, maybe I don't think it's okay for certain applications to use persistant storage for temporary files.
>>300892 >HDD for /home Never thought of this, but won't steam games generally run slower since the default installation places it there? I will also never understand why Valve makes the steam folder hidden.
Has anybody have issues with your mouse sensitivity not showing up in the settings? There is acceleration, but I usually disable that shit by default. Can't find much good documentation on editing it manually via terminali also have a hard time finding info.
>>302150 Forgot to mention, using xfce4
>>302144 Worst that could happen is that games would load slower but it shouldn't really affect performance. Either way you can solve this by symlinking the steam folder somewhere else. I've done that with wine folders (/home is on an SSD and everything that needs a lot of space is offloaded to HDDs).
>>302150 asfaik xfce4 sucks when handling libinput drivers. so theres no good GUI for mouse settings on xfce4 Good news, since it's all going through xorg and libinput you can easily do it from terminal. Try this script: for i in {0..99}; do xinput set-prop $i "libinput Accel Profile Enabled" 0 1 &> /dev/null done All it does it iterate through xinput devices and turns off anything related to acceleration.
>>302144 you can change the steam library DIR in steam's download settings.
>>302144 I forgot about that since I have my steam library on a separate drive. I use an nvme for games since if the nvme dies it's not like I can't just download them again. Unless it's cracked games, those I keep backed up.
>>302144 >but won't steam games generally run slower since the default installation places it there? While yes you will have faster performance, even when sharing with general disk access Having your videogames not throttled by sharing disk bandwidth with system processes is a pretty noticable boost for performance
>>302159 Thanks anon.
>>302159 >>302183 Tried the script, only put my mouse to a slog. Which of of these can I edit to increase the sensitivity? Tried changing some of the numbers around but with the exact same results.
>>302389 >only put my mouse to a slog turning off acceleration tends to do that. I probably could make a script but i'm lazy so here: # find device and it's id xinput --list --short # list properties to confirm it has "Device Accel Constant Deceleration" xinput --list-props <device id> # change Device Accel Constant Deceleration xinput --set-prop <device id> "Device Accel Constant Deceleration" 1.5 # (1.5 or desired sensitivity. sens is done using a floating point value) actually a script would probably look like: # usage: uses decimal value from stdin. E.G: # larry ~ $ ./script.sh 1.5 # will give you a sens of 1.5 for i in {0..99}; do xinput set-prop $i "Device Accel Constant Deceleration" $1 &> /dev/null done That'd probably work. I don't use xorg so i can't test it. An easier method of changing sensitivity would probably be to just change the settings directly on the mouse's firmware if you got a mouse that can do it. Example, logitech mice are very usable on linux as someone made a helpful application to do this for you called solaar. you can change keymaps, dpi and mouse sensitivity through it. For other vendors you'd probably either need to use their windows driver to change the firmware or do the hard method of digging through the /dev/ files and attempting to write values to them (not recommended).
>>302713 >$1 &> /dev/null shit, typo in the script. oh well. should look like $1 1 &> /dev/null but i fugged it.
>>302713 Funnily enough I do have a Logitech MMO mouse, so that program will be nice. I need to keep studying shell scripting. >>302719 Thanks
(37.21 KB 616x353 capsule_616x353.jpg)
Is the Steam controller any good? If so can it be used as a regular controller on Linux without having to make a Steam account to use it?
>>303105 There was a software for it called SC-controller but i'm not sure the dev still works on it. Also steam controller has lots of issues on wayland so if you use that it may not do everything properly.
>dolphin-emu native >OpenGL runs at sub-40% native speed on AyyMD, Vulkan runs fine and is the only practically usable backend >dolphin core on retroarch >Vulkan crashes but the OGL backend runs at almost the same speed as the native VK one How?
>>303105 Other than what >>303112 mentioned (I think it got abandoned some time ago), it definitely works as mouse on stock Fedora, but I've never tried more than that. I might be misremembering but I think there was a way to setup controller behaviour outside steam inside the steam client. On windows it's decent at what it tried to do but for most PC games it's worse than mouse and keyboard, especially where aiming is involved. There's also a learning curve to using it and you should definitely look up controller configurations other people use. Gyro+"trackball" aim is still far superior to joystick aim. Using the trackpads as substitute for time critical input buttons sucks and using them as joysticks is also pretty bad. The grip might also be uncomfortable depending on your hands and how you play. It might be worth a buy but you definitely need to consider the tradeoffs.
(775.80 KB 640x360 wendel_laugh_apple.webm)
>>303105 It's a tinkerer's gamepad. For use on odd games that are generally not meant to be played with a controller or other weird shit where you want to play around and customise heavily. Outside of that, if you just want a reliable, well-designed no-nonsense gamepad to use under steam then you should just go with one of the console controllers. I use the updated PS4 pad and it's fine for DMC and Tekken. >>302159 This makes my mouse_accel_off script look like a pile of shit. Good idea to have a script that sets your monitor to the res/rotation/location/Hz that you're used to. Just in case some old wine games fuck with your screen and you want an easy to execute script to fix it all. Here's mine as an example #!/bin/sh xrandr --output DisplayPort-0 --mode 2560x1440 --rate 143 --output DVI-D-0 --mode 1440x900 --rate 75.02 --rotate right --left-of DisplayPort-0


Quick Reply
Extra
Delete
Report

no cookies?